Skip to content

Today's NBA Games

Wednesday, March 25, 2026

Final/OT ESPN
Hawks 41-32
130 129
Pistons 52-20
Final Spectrum Sports Net, Spectrum Sports Net +, NBA League Pass
Lakers 47-26
137 130
Pacers 16-57
Final CHSN, NBA League Pass
Bulls 29-43
137 157
76ers 40-33
Final NBC Sports BO
Thunder 57-16
109 119
Celtics 48-24
Final FanDuel SN Sun, NBA League Pass
Heat 39-34
120 103
Cavaliers 45-28
Final FanDuel SN SE
Spurs 55-18
123 98
Grizzlies 24-48
Final MNMT, NBA League Pass
Wizards 17-55
133 110
Jazz 21-52
Final/OT ESPN
Rockets 43-29
108 110
Timberwolves 45-28
Final KFAA-TV, Mavs.com, NBA League Pass
Mavericks 23-50
135 142
Nuggets 46-28
Final NBC Sports BA
Nets 17-56
106 109
Warriors 35-38
Final KUNP 16, BlazerVision
Bucks 29-43
99 130
Trail Blazers 37-37
Final TSN, NBA League Pass
Raptors 40-32
94 119
Clippers 37-36

Strong model edge (6%+)   Mild lean (3–5%)   Market-aligned · Edge = TankOdds win % vs market-implied win % from spread

Last updated 4:11 AM EDT ET · Scores update on page refresh

Today's NBA Odds — Spreads, Totals, and Win Probability

Every game card on this page shows the current point spread and over/under pulled from ESPN's live scoreboard feed — the same lines sportsbooks use as a baseline. Run the sim on any matchup and your browser runs 10,000 independent trials in a few seconds. What you get back: win probability for each team, the projected final score, cover percentage for the favored side, and a reading on whether the model's estimate diverges from what the spread implies.

Reading NBA point spreads

The spread is the margin the oddsmakers expect between the two teams. A favorite at -7.5 has to win by 8 or more to cover. The underdog at +7.5 covers if they win outright or lose by 7 or fewer. Cover % in the simulation shows how often the favored team beat the spread across all 10,000 trials. Near 50% means the model sees the current line as fairly priced. Significantly above or below suggests the model reads the matchup differently than the market does.

Understanding NBA over/unders

The over/under is the projected combined total for both teams. The Over % in the sim is how often the two teams' simulated scores added up to more than that number. Like cover %, it's a probability estimate — close to 50% means the model agrees with the total as priced, further from 50% means it sees a lean in one direction.

What the edge vs. market row means

The colored row at the bottom of each sim result compares TankOdds win probability to the win probability implied directly by the spread, using the same normal distribution for both. A green edge (6%+) means the model's win % for that team is at least 6 points above what the spread is pricing in — a real divergence, not noise. Amber (3–5%) is a smaller lean. Gray means the model and the market are in rough agreement. This is a diagnostic, not a pick. The model doesn't know about injuries, rest days, or lineup changes.

How the TankOdds Simulation Works

When you click Run Simulation, your browser runs 10,000 independent trials for that game. No server, no delay. Each trial generates a score for both teams by sampling from a normal distribution built from the spread and over/under. The favorite's scoring mean is set higher, the underdog's lower, and both teams get a standard deviation of 11 points to reflect the variance you actually see in NBA games.

What each number means

Win % is how often that team won across all 10,000 trials. If it says 65%, that team came out on top in 6,500 simulations. It's a probability estimate based on the current odds, not a prediction.

Projected score is the average final score across all trials. Think of it as the center of the scoring range the model sees, not a specific call on what will happen.

Cover % is how often the favored team beat the spread. Near 50% means the model thinks the line is fair. Well above or below means it sees the line as off.

Over % works the same way for the total. Close to 50% means the model agrees with the over/under as priced.

The edge color explained

The colored row at the bottom compares our win probability against what the spread itself implies. Both calculations use the same normal distribution with a standard deviation of 11, so it's a clean apples-to-apples comparison.

Green (6%+ gap) means our win probability for that team is at least 6 points above what the spread implies. That's a real divergence worth paying attention to.

Amber (3-5% gap) is a smaller lean. The model favors one side but not by a wide margin.

Gray means the model and the market are basically in agreement.

What this tool is and isn't

This is a statistical simulation for educational and entertainment purposes. It's not gambling advice. The model only sees the spread and over/under. It has no idea about injuries, rest days, lineup changes, or anything else a real oddsmaker factors in, so treat it accordingly.

When the model "finds value," that means its win probability differs from what the spread implies. It doesn't mean you should bet on it. Sports betting carries real financial risk. TankOdds does not accept wagers and does not endorse any sportsbook.

Game times are in your local timezone. Scores and odds update on page refresh. Must be 21+ to access any linked sportsbook. Sports betting isn't legal everywhere, so know your state's laws. Gambling problem? Call or text 1-800-522-4700. Responsible Gambling resources.